Decoy Cells versus Plasma Real-Time Polymerase Chain Reaction for the Detection of Polyomaviruses in Renal Transplant Patients: A Single Institutional Experience.

BACKGROUND AND AIMS: Microscopic detection of decoy cells is routinely performed in urine samples from renal transplant patients for the evaluation of polyomaviruses. However, they are scanty papers evaluated the diagnostic accuracy of decoy cells in urine samples. The aim of this study is to evaluate the diagnostic accuracy of decoy cells in urine samples and compare with plasma real-time polymerase chain reaction (RT-PCR) as a gold standard method. In addition, to compare the findings of this study with other similar studies.
METHODS: A retrospective study over a period of four years from January 2014 to December 2017 was performed. A total of 89 urine samples from renal transplant patients were assessed for the presence of polyomaviruses and compared with plasma RT-PCR. The sensitivity, specificity, accuracy, positive predictive value (PPV) and negative predictive value (NPV) were measured.
RESULTS: There were 29 males and 18 females. The mean patient age was 40.3 years. The sensitivity, specificity, accuracy, PPV and NPV were 86.6%, 67.5%, 70.7%, 35.1% and 96.1%, respectively. Other similar studies reported a sensitivity of 41.9-84.6%, specificity of 65.8-100% and accuracy of 69.9-82%.
CONCLUSION: The findings of this study show that the detection of decoy cells in urine samples is a sensitive screening method for polyomaviruses. The findings of this study are compatible with other similar studies. Copyright:

INTRODUCTION

BK and JC polyomaviruses are the major pathogenic viruses in humans.[1] They are naturally found in healthy individuals without causing harms as about 70 – 90% of adults have antibodies against them.[2] Following infection, these viruses remain latent, mainly in the renourinary tract, such as in the bladder transitional cell layer, renal pelvis, ureters and in tubular epithelial cells of the kidney. The activation of these viruses is increased during strong immunosuppression therapy usually after renal transplantation.[3] This activation may lead to polyomavirus associated nephropathy (PVAN).[4] PVAN is one of the major causes for allograft failure.[5] PVAN is mainly caused by BK polyomaviruses, and to a lesser extent, by JC polyomaviruses.[6] The incidence of PVAN ranges from 1% to 10% in renal transplant recipients.[7]

Reactivation of polyomaviruses leads to the shedding of infected cells into urine, these cells called Decoy cells. These cells are easily identified in urine smears using routine Papanicolaou (PAP) or hematoxylin and eosin stains (H and E). Despite the use of urine cytology as a routine work, they are scanty papers evaluated the diagnostic accuracy of decoy cells in urine. The aim of this retrospective study was to compare the sensitivity, specificity, accuracy, positive predictive value and negative predictive value for routine decoy cells in urine samples with the plasma real-time polymerase chain reaction (RT-PCR) as a gold standard method. In addition, the findings of this study were compared with other similar studies.

METHODS

A retrospective study over a period of four years from January 2014 to December 2017 was performed in the Department of Pathology, Sultan Qaboos University Hospital, Sultanate of Oman. The population was all renal transplant cases diagnosed for polyomavirus by both methods: urine cytology and plasma RT-PCR. This study was ethically approved by the Medical Research Committee and Ethics Committee at the College of Medicine and Health Sciences, Sultan Qaboos University, Sultanate of Oman. (MREC # 1538).

Urine samples were macroscopically evaluated for the volume, color and appearance. The samples were then centrifuged at 1800 rpm/5 min (centrifuge 5702, Eppendrof, Hamburg, Germany) and the supernatant was removed. Two microscopic slides were prepared for each urine sample and stained with PAP and H and E. Finally, the slides were screened by cytotechnologist and double checked by cytopathologist. Actually, the reporting terminology in cytology laboratory for positive cases of polyomaviruses were: adequate, negative for malignant cells and presence of decoy cells.

Plasma samples in EDTA preservative tubes were prepared for referral transportation. The primary reason for referral was because of the few number of cases. In abroad laboratory, polyomavirus nucleic acids were extracted using a Wizard TM Genomic DNA Purification kit (Promega Corporation. Madison, WI, USA). RT-PCR was applied using artus BK Virus RG PCR (QIAGEN GmbH, Hilden, Germany) which consists of a ready-to-use system for detection of BK virus DNA on Capillary based instrument (Rotor-Gene Q MDx/Rotor-Gene Q software version 1.7.94). The instrument consists of four program steps: denaturation, when DNA was melted and polymerase enzyme was activated. Annealing, when the reaction temperature was lowered allowing annealing of the primers to each of the single stranded DNA templates. Extension/elongation, the DNA polymerase synthesized new DNA strands. Cooling of the instrument at the end of each run. Finally, the results were analyzed using the instrument's software by measuring viral load for each sample.

The data were analyzed using the Statistical Package for the Social Sciences (SPSS) software version 23 (SPSS Inc., Chicago, IL, USA). Sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), false positive and false negative values were used for analysis of urine cytology compared to the plasma RT-PCR as a gold standard method.

RESULTS

A total of 47 renal transplant patients were identified. There were 29 males and 18 females. The mean patient age was 40.3 years (age range being 16-68 years). A total of 89 urine cytology analysis were compatible with plasma quantitative RT-PCR tests which were performed in samples within three weeks. In contrast, the rest of the cases that had these two tests but the samples were collected in different dates were excluded from the study. Among 89 cases, the sensitivity, specificity, accuracy, PPV and NPV were 86.6%, 67.5%, 70.7%, 35.1% and 96.1%, respectively. Other parameters are shown in Table 1. An example of decoy cells in urine samples is shown in Figure 1. The findings of this study are comparable with other similar studies [Table 2].

Table 1

Comparison between urine cytology and RT-PCR in the diagnosis of polyomaviruses in renal transplant patients

Plasma RT-PCR
		Positive	Negative	Total
Urine cytology (PAP smear)	Positive	13	24	37
	Negative	2	50	52
	Total	15	74	89

Figure 1

Decoy cells in urine samples. (a) Papanicolaou method, ×600. (b) haematoxylin and eosin method, ×600

Table 2

Comparison between urine cytology and RT-PCR in the diagnosis of polyomaviruses in the current study and other similar studies

Studies	No of samples	Sensitivity %	Specificity %	Accuracy %	PPV %	NPV %
Current study	89	86.6	67.5	70.7	35.1	96.1
Xing et al.[8]	32	75	75	75	75	75
Zalona et al.[9]	104	61.1	74.4	72.1	33	90.1
Ranzi et al.[10]	442	84.6	68.5	69.9	21.2	97.8
Randhawa et al.[11]	100	41.9	100	82	100	82.8

PPV: positive predictive value, NPV: negative predictive value.

DISCUSSION

The evaluation of urine cytology, via the detection of decoy cells in urine samples, in the diagnosis of PVAN is important for a better clinical management. It is widely used for the diagnosis of polyomaviruses.[12] We do emphasis, as with others, that early screening for polyomaviruses in renal transplant patients may reduce the severity of PVAN.[131415] Until know, there is no effective treatment for polyomaviruses. The only available treatment method is to reduce the doses of immunosuppressive drugs with a close follow up. Thus, will improve and prolong graft survival.[1617]

In this study, urine cytology shows an 86.6% sensitivity in the detection of polyomaviruses. This finding is higher than other similar studies.[8911] However, it is less than other study, which reported a sensitivity of 96.7%, when they compared urine cytology with the histopathological diagnosis of renal biopsy.[18]

Urine cytology involves noninvasive method, cheaper than other methods such as renal biopsy, urine or plasma RT-PCR and electron microscope, easier to perform, and can be performed in a basic laboratory.[19] Cytologically, decoy cells are usually identified by condensation of chromatin and large granular intranuclear inclusion bodies with a ground glass nucleus appearance. In addition, four different variants of decoy cells have been identified.[20] Unfortunately, urine cytology cannot differentiate BK from JC polyomaviruses. Decoy cells could be the result of infection with BK, JC, or to a lesser extent, adenoviruses.[21] However, it is very rare that JC polyomavirus causes PVAN.[2223] Although identification of decoy cells in urine smears is relatively easy, accurate diagnosis by cytoscreeners and cytopathologists is important to distinguish decoy cells from other viral infection, reactive atypia, urothelial dysplasia and malignant tumour cells.[24] The major benefit of using RT-PCR in the diagnosis of polyomaviruses is that it can detect very low levels of the virus replication.[25] However, RT- PCR analysis is expensive, difficult and contamination might be a problem.[26] In fact, RT-PCR is only carried out in patients with persistent viruvia, as the RT-PCR result does not contribute in management decisions.[15] Positive plasma RT PCR might indicate an increased risk for PVAN, but renal biopsy is essential for diagnosis. Histopathological diagnosis of renal biopsy samples is the definitive diagnosis of PVAN and more accurate than RT-PCR.[27]

In the literature review, few studies compared plasma and urine PCR in detecting polyomaviruses in renal transplant patients. For example, Chung and colleagues demonstrated that both plasma and urine PCR showed 100% sensitivity and 100% NPV. However, specificity (91.8%) and PPV (27.3%) for urine PCR were less compared with plasma PCR, which showed specificity of 97.4% and PPV of 54.5%.[28] Other two similar studies in Italy and USA revealed almost the same findings.[2930] It is important to note that the presence of polyomaviruses in renal transplant patients is first detected in urine then in plasma few weeks later and eventually in the renal tissue.[31] This sequential order (viruria-viremia-nephritis) of virus replication in renal transplant patients fits with the diagnostic tests of urine cytology or urine PCR, plasma PCR and renal biopsy.[32] Urine PCR is more specific than the detection of decoy cells in urine samples of renal transplant patients.[33] Several researchers reported that the duration from viruria to viremia is one to three months, and from viremia to PVAN is one to twelve weeks.[132834] Thus, for an early detection of viruria, urine samples are more appropriate. Although there is no an international cutoffs for viruria and viremia, viral load of >104 copies/ml and >107 copies/ml have been strongly linked with nephropathy in renal transplant patients, respectively.[1430]

In this study, we report a low specificity (67.5%). This finding disagrees with other reported studies, which reported higher specificity.[891118] There are many factors that might contribute to the low specificity such as delayed delivery of urine samples, improper sample preparation and improper staining.[13] In addition, high false positive cases may be due to the presence of nonspecific viral cytopathic changes, cellular degeneration and morphologic mimic of high grade urothelial carcinoma cells.[8]

The negative predictive value (96.1%) of this study is higher than other studies[8911] and very close to the other study.[10] However, the positive predictive value (35.1%) is less than other reported studies.[81118] In addition, this study shows an accuracy rate of 70.7%, which is in line with other similar studies.[8910] Other study which scored a high accuracy rate (97%), their comparison was between urine cytology and histopathological diagnosis of renal biopsy samples and not with RT-PCR as with others.[18]

Several limitations of our study are worth noting. Firstly, this study is a retrospective study. Secondly, plasma samples for RT-PCR were processed abroad. Thirdly, urine RT-PCR was not measured as polyomaviruses remain positive in urine longer than in blood.[35] Finally, the absence of histopathological diagnosis for renal biopsies.

In conclusion, the findings of this study show that the microscopic detection of decoy cells in urine samples is a sensitive screening method for polyomaviruses. The findings of this study are compatible with other similar studies.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.